Skip to main content
Log in

A characterization of the Wishart exponential families by an invariance property

  • Published:
Journal of Theoretical Probability Aims and scope Submit manuscript

Abstract

E is the space of real symmetric (d, d) matrices, andS and\(\bar S\) are the subsets ofE of positive definite and semipositive-definite matrices. Let there be ap in

$$\Lambda = \left\{ {\frac{1}{2},1,\frac{3}{2}, \ldots \frac{{d - 1}}{2}} \right\} \cup \left] {\frac{{d - 1}}{2}, + \infty } \right[$$

The Wishart natural exponential family with parameterp is a set of probability distributions on\(\bar S\) defined by

$$F_p = \{ \exp [ - \tfrac{1}{2}Tr(\Gamma x)](det\Gamma )^p \mu _p (dx);\Gamma \in S\} $$

where μp is a suitable measure on\(\bar S\). LetGL(ℝd) be the subset ofE of invertible matrices. Fora inGL(ℝd), define the automorphismg a ofE byg a(x)=t axa, whereta is the transpose ofa. The aim of this paper is to show that a natural exponential familyF onE is invariant byg a for alla inGL(ℝd) if and only if there existsp in Λ such that eitherF=F p, orF is the image ofF p byx↦−x. (Theorem).

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Similar content being viewed by others

References

  1. Barndorff-Nielsen, O. (1978).Information and Exponential Families in Statistical Theory. Wiley, New York.

    Google Scholar 

  2. Efron, B. (1978). The geometry of exponential families.Ann. Stat. 6, 362–376.

    Google Scholar 

  3. Letac, G., and Mora, M. (1987). Natural exponential families with cubic variances. To appear inAnn. Stat.

  4. Anderson, T. W. (1958).An Introduction to Multivariate Statistical Analysis, Wiley, New York.

    Google Scholar 

  5. Muirhead, R. J. (1982).Aspects of Multivariate Statistical Theory. Wiley, New York.

    Google Scholar 

  6. Gindikin, S. G. (1975). Invariant generalized functions in homogeneous domains.Funct. Anal. Appl. 9, 50–52.

    Google Scholar 

  7. Artzner, Ph. (1970). Sur la décomposition de la loi du 85-1 à plusieurs dimensions.C. R. Acad. Sci. Paris. Sér. A 271, 372–375.

    Google Scholar 

  8. Lévy, P. (1948). The arithmetical character of the Wishart distribution.Proc. Cambridge Phil. Soc. 44, 295–297.

    Google Scholar 

  9. Morris, C. M. (1982). Natural exponential families with quadratic variance functions.Ann. Stat. 10, 65–80.

    Google Scholar 

  10. Meimné, R., and Testard, F. (1986).Introduction à la Théorie des Groupes de Lie Classiques. Hermann, Paris.

    Google Scholar 

  11. Gindikin, S. G. (1964). Analysis on homogeneous domains.Russ. Math. Surv. 19, 1–89.

    Google Scholar 

  12. Vinberg, E. B. (1963). Theory of convex homogeneous cones.Trans. Moscow Math. Soc. 12. (A.M.S. Translations).

  13. Artzner, Ph., and Fourt, G. (1974). Lois gamma et bêta sur un cône convexe homogène. Applications en Analyse multivariée.C. R. Acad. Sci. Paris, Sér. A 278, 293–295.

    Google Scholar 

  14. Eriksen, P. S. (1985). (k, 1) Exponential transformation models.Scand. J. Stat. Theoret. Appl. II, 129–145.

    Google Scholar 

  15. Watson, G. S. (1984).Statistics on Spheres, Wiley, New York.

    Google Scholar 

  16. Jensen, J. L. (1981). On the hyperboloid distribution.Scand. J. Stat. 8, 193–206.

    Google Scholar 

Download references

Author information

Authors and Affiliations

Authors

Rights and permissions

Reprints and permissions

About this article

Cite this article

Letac, G. A characterization of the Wishart exponential families by an invariance property. J Theor Probab 2, 71–86 (1989). https://doi.org/10.1007/BF01048270

Download citation

  • Received:

  • Issue Date:

  • DOI: https://doi.org/10.1007/BF01048270

Key Words

Navigation